Cutaneous angiosarcoma is a rare cancer for which effective treatment has not been developed sufficiently. Still, because it often occurs in elderly people, the number of patients is increasing due to the aging population. Cutaneous angiosarcoma is difficult to completely remove by surgery, and recurrence and metastasis after surgery are not uncommon. Therefore, chemotherapy, radiation therapy, and a combination of these are currently widely used as treatments. Traditionally, the anticancer drug used for cutaneous angiosarcoma was mainly doxorubicin. In recent years, it has been reported that cancer shrank in 18% of patients after two months of paclitaxel administration in a clinical trial. Therefore, paclitaxel has become more commonly used for cutaneous angiosarcoma. Other options include anthracycline anticancer drugs and gemcitabine. However, even with these anticancer drugs (and radiation therapy), cutaneous angiosarcoma progresses quickly, and some reports have said that the 5-year survival rate is 9%. This study is planned to develop a safer and more effective treatment for cutaneous angiosarcoma and will include 38 participants. In this study, eligible participants will receive combination chemotherapy with 200 mg of pembrolizumab (injection liquid, once every 3 weeks) and 20 mg of lenvatinib (capsule, once daily) for up to approximately two years as protocol treatment unless the criteria for termination meet. Before, during, and after the protocol treatment, participants will undergo many examinations and evaluations, including blood tests, urine tests, and imaging tests (e.g., x-ray, CT scan, or MRI) to assess the safety and efficacy of the protocol treatment.

This is a multi-center, single-arm, investigator-initiated phase II clinical trial to evaluate the efficacy and safety of pembrolizumab plus lenvatinib therapy in patients with unresectable cutaneous angiosarcoma (WHO Classification of Skin Tumors 4th edition).

In Japan, the first-line treatment for cutaneous angiosarcoma is surgical resection with radiotherapy for patients with limited and completely resectable lesions and chemotherapy alone or with radiotherapy for patients with unresectable lesions. Paclitaxel is widely used for chemotherapy based on the ANGIOTAX trial. However, the prognosis of Japanese patients with cutaneous angiosarcoma is still poor, with a mean survival time of 19.52 months and a 5-year survival of 9%. The development of a new therapeutic regimen is an urgent task to improve the prognosis of such patients; however, company-sponsored clinical trials for this disease are not actively conducted since cutaneous angiosarcoma is a rare cancer. For the above reasons, conducting this study as an investigator-initiated clinical trial is important. The clinical hypothesis of this study is "the efficacy of pembrolizumab plus lenvatinib combination therapy exceeds ANGIOTAX trial in the primary endpoint of response rate, and the efficacy and safety present expecting results in the secondary endpoints." According to an epidemiological survey of skin cancer based on the National Cancer Registry, the total number of cutaneous angiosarcoma registered in 2016 and 2017 was 993, with an average annual registration of 497 cases. Cutaneous angiosarcoma is a very rare cancer, and it is impossible to secure enough participants to conduct a randomized controlled clinical trial. Therefore, this study is designed as a single-arm study. The result of a phase II study (Alliance A091902) of cabozantinib + nivolumab for advanced angiosarcoma previously treated with a taxane-based regimen was reported at ASCO 2023, and the overall response rate for cutaneous angiosarcoma (N=13) was 54% among the 22 cases registered to the trial. It is expected that a high response rate will also be obtained in this study regarding a similar regimen. Since this is a single-arm study for rare cancer, eligible patients who give consent are expected to be enrolled almost comprehensively, so selection bias is unlikely to occur. The primary endpoint of this study is the response rate, a common endpoint in phase II trials. Since the response rate in the ANGIOTAX trial was 18%, the threshold response rate for this trial will be set at 18%, the expected response rate at 35%, the significance level at 0.05, and the statistical power at 0.70. According to this study's threshold response rate, expected response rate, significance level, and statistical power, the number of participants required for analysis will be 35. The planned number of participants to be enrolled was set at 38, anticipating some dropouts. In the ANGIOTAX trial, 11 participants among the 30 enrolled were previously treated with chemotherapy, and 19 were untreated. To ensure comparability with the trial, the maximum number of previously treated and untreated participants to be enrolled in this study is set at 13 and 25, respectively. The ANGIOTAX study reported that the progression-free survival at the second cycle (10 of 13 previously treated, 13 of 14 untreated) and median survival time (p=0.37 by log-rank test) were similar between previously treated and untreated patients. If at least three cases of complete response (CR) or partial response (PR) are not confirmed at the 12-week efficacy evaluation in the first 20 participants out of the planned 38, the protocol treatment will be considered not effective enough, and early termination of the study will be discussed. Pembrolizumab and lenvatinib are unapproved drugs for cutaneous angiosarcoma in Japan and cannot be used in the medical insurance scheme. Participants enrolled in this study will have the opportunity to receive treatment with the off-label drugs of pembrolizumab and lenvatinib. Since the efficacy and safety of pembrolizumab and lenvatinib combination therapy for cutaneous angiosarcoma is unclear, it is unknown whether clinical benefits such as tumor shrinkage and prognosis extension will be obtained. However, although the cancer type is different, it has been reported that the combination of lenvatinib and an anti-PD-1 antibody showed enhanced tumor growth inhibition in renal cell carcinoma cell lines compared to the single administration of lenvatinib or an anti-PD-1 antibody. In addition, it has been suggested that lenvatinib increases the proportion of activated cytotoxic T cells in lymph nodes and reduces the proportion of TAM, an immunosuppressive cell, in tumor tissue. Since lenvatinib with an anti-PD-1 antibody may enhance the anti-tumor immune response, it is anticipated that the combination of pembrolizumab and lenvatinib may further enhance the therapeutic effect in cutaneous angiosarcoma. Pembrolizumab and lenvatinib may cause acute or late toxicities, including immune-related adverse events, which are expected disadvantages of participating in this study. Inclusion/exclusion criteria and treatment modification criteria were carefully considered to minimize the impact of such toxicities. The pembrolizumab and lenvatinib combination regimen was determined based on the phase 1b part of the phase 1b/2 trial 111/KEYNOTE-146. 111/KEYNOTE-146 was primarily aimed at determining the maximum tolerated dose (MTD) and phase II recommended dose (RP2D) of lenvatinib in combination with pembrolizumab 200 mg IV Q3W (treatment dose for all approved indications at the beginning of 111/KEYNOTE-146 trial). The initial dose of lenvatinib was 24 mg QD (the starting dose currently approved as monotherapy for differentiated thyroid carcinoma) and was administered orally. The Phase 1b part of the trial enrolled 13 participants (pembrolizumab 200 mg IV Q3W plus lenvatinib 24 mg QD: n = 3; pembrolizumab 200 mg plus lenvatinib 20 mg QD: n = 10). Among the subjects, 8 had renal cell carcinoma, two non-small cell lung cancer, two endometrial cancer, and one melanoma. Two dose-limiting toxicities (DLT) occurred with pembrolizumab 200 mg Q3W plus lenvatinib 24 mg QD (one participant experienced Grade 3 joint pain and the other Grade 3 exhaustion). No DLT was observed with pembrolizumab 200 mg Q3W plus lenvatinib 20 mg QD. Accordingly, the RP2D of combination therapy was determined as pembrolizumab 200 mg Q3W plus lenvatinib 20 mg QD. The RP2D was evaluated for efficacy and safety in the phase II study. Based on the encouraging antitumor effect and a tolerable safety profile found in multiple tumor types in the 111/KEYNOTE-146 trial, two Phase III trials, 309/KEYNOTE-775, and 307/KEYNOTE-581 have been initiated, and drug combinations have been approved in multiple regions for the treatment of these tumor types. One course of this study is set at 21 days. The length of each course will not be extended or shortened. After confirmation that the participant does not fall under any of the criteria for termination or suspension, 200 mg of pembrolizumab is administered intravenously on day 1 of each course in outpatient settings. The intravenous injection lasts 30 minutes, with an allowance of -5 to +10 minutes. If pembrolizumab cannot be administered (e.g., because the participant meets suspension criteria), its administration will be reconsidered on day 1 of the next course. Participants take 20 mg of lenvatinib capsule orally, once daily (from day 1 to 21 of every course). On the days when the participant receives both pembrolizumab and lenvatinib, either drug can be administered first. Lenvatinib should be taken at the same time every day as possible, regardless of whether it's taken with or without food or examination, before/after/between meals. This study's dose of pembrolizumab is only one level, 200 mg. Lenvatinib is started at 20 mg QD, which will be gradually tapered to 14 mg, 10 mg, and 8 mg if the participant falls under any of the dose reduction criteria. The minimum dose of lenvatinib is 8 mg QD, and a lower dose should not be accepted. Completion of protocol treatment is defined as completing 35 courses of pembrolizumab and/or lenvatinib. Tumor assessment is performed during the screening period, every 6 weeks from the initiation of protocol study to 24 weeks, and every 12 weeks afterward, by CT, MRI, or visual inspection (colored photograph). The modality and imaging conditions used at screening should be used for the consecutive assessments. Imaging tests or visual inspections should be continued until the overall response turns to PD or the post-study treatment starts. If the overall response of CR or PR was first detected, the next imaging test or visual inspection for confirmation should be performed 4 weeks after that day or later. An interim analysis will be conducted during the registration period to evaluate whether the response rate is inferior to that expected and determine if enrollment should be ceased (and the study be terminated due to ineffectiveness). The interim analysis will be conducted in accordance with a SWOG two-stage design. In the 1st stage, the enrollment is suspended when the 20th patient is enrolled, and the data, such as response evaluation used for the interim analysis, will be locked for around 12 weeks after the enrollment of the 20th patient. Even if the 20th patient is not evaluated 12 weeks after the start of treatment, the timing of the data lock will not be changed. The best overall response by central review will be used to calculate the response rate in the interim analysis. The FAS at the time of the interim analysis is the population analyzed. After fixation of the data on the best overall response by central review 6 months after the end of the registration period, an analysis mainly on the primary endpoint will be performed (the main analysis). The population analyzed will be FAS. The main analysis aims to determine whether the pembrolizumab plus lenvatinib therapy is sufficiently effective using the primary endpoint of overall response rate by central review. If the null hypothesis "the true response rate is below the threshold response rate (18%)" is not rejected, the primary endpoint is not significantly different from てthe ANGIOTAX trial, and the protocol treatment will not be considered superior. If the null hypothesis is rejected, the primary endpoint significantly differs from the ANGIOTAX trial, and the protocol treatment is considered superior after further consideration of the secondary endpoints. In this case, the application for approval of the combination therapy of pembrolizumab and lenvatinib to expand indications for treating unresectable cutaneous angiosarcoma will be considered.

Inclusion Criteria:

1. Histologically diagnosed with cutaneous angiosarcoma.
2. Unresectable primary or metastatic disease.
3. No spinal metastases that require radiotherapy or surgical intervention.
4. No pericardial effusion, pleural effusion, or ascites that require treatment.
5. Age at enrollment: 18 years or older and 85 years or younger.
6. The most recent performance status score (ECOG) within 14 days prior to enrollment is 0 or 1.
7. A contrast-enhanced CT/MRI (head, neck, chest, abdomen, pelvis: slice thickness 5 mm or less) performed within 14 days prior to enrollment or have at least one measurable lesion that can be confirmed to be present on the surface of the body by visual inspection (for patients with contrast allergy or renal function deterioration, plain CT is allowed).
8. Cohort A (untreated patients): Cutaneous angiosarcoma without prior therapy. Cohort B (previously treated patients): Cutaneous angiosarcoma with prior therapy.
9. Have adequately controlled BP with or without antihypertensive medications (of 2 or less counted by components), defined as BP <=150/90 mmHg with no change in antihypertensive medications within 7 days prior to enrollment.
10. Wounds (e.g., those after a surgical procedure or trauma) have healed at enrollment.
11. Did not have major surgery within 21 days prior to the first dose of study interventions. Note: Adequate wound healing after major surgery must be assessed clinically, independent of time elapsed for eligibility.
12. Laboratory tests performed within 14 days before the enrollment date meet the following criteria: (1) to (10). However, patients should not receive granulocyte colony-stimulating factor (G-CSF) or blood transfusions within 14 days before the day of blood collection.

(1) Neutrophil count >=1,500/mm3 (2) Platelet count >=10×104/mm3 (3) Hemoglobin >=9.0 g/dL (4) AST <=100 U/L (5) ALT<=100 U/L (6) Total bilirubin <=1.5 mg/dL (7) Serum creatinine <=1.5 mg/dL (8) Creatinine/clearance >=30 mL/min (Cockcroft-Gault). (Even if the calculated value is less than 30 mL/min, it can be registered if the value is more than 30 mL/min by the 24-hour urine collection method).

(9) Urine protein <1 g/24 hours. Note: Participants with proteinuria >=2+ on urine dipstick testing (urinalysis) will undergo 24-hour urine collection for quantitative assessment of proteinuria.

(10) PT-INR <=1.5 13) Male patients agree to use highly effective contraception and not donate sperm during treatment and for at least 120 days after the last dose of the investigational drugs.

Female patients agree to use contraception methods, are not pregnant or breastfeeding, and at least one of the following conditions applies:

• Is not a woman of childbearing potential (WOCBP).

• Is a WOCBP tested negative by pregnancy test within 14 days prior to enrollment, and using a contraceptive method that is highly effective, with low user dependency, or be abstinent from heterosexual intercourse as their preferred and usual lifestyle during the intervention period and for at least 120 days post pembrolizumab or 30 days post lenvatinib, whichever occurs last. A WOCBP must have a negative result of a highly sensitive pregnancy test within 24 hours before the first dose of study intervention.

  14) Twenty-eight days have passed since the last day of prior chemotherapy if applicable, and the patient has recovered from all toxicities except alopecia and blood toxicity (below Grade 1 or to baseline).

  15) Written informed consent for study participation has been obtained from the patient.

Exclusion Criteria:

1. Persons previously treated with anti-PD-1, anti-PD-L1, anti-PD-L2 agents, or agents targeting other co-inhibitory T-cell receptors (e.g., CTLA-4, OX-40, CD137).
2. Individuals previously treated with lenvatinib or another angiogenesis inhibitor, including treatment for other cancer types.
3. Patients who are likely to be cured with chemoradiotherapy, at the discretion of the treating physician.
4. Patients who received radiation therapy within 14 days prior to enrollment or patients with radiation therapy-related toxicities at enrollment.
5. Patients positive for either HIV antibody, HBs antigen, or HCV antibody.
6. Patients of HBs antigen-negative, HBs antibody, or HBc antibody-positive, and HBV-DNA assay-positive.
7. Patients receiving live or live attenuated vaccines within 30 days of enrollment.
8. Patients who received other investigational products or used investigational medical devices within 28 days prior to enrollment.
9. Patients diagnosed as being in an immunocompromised state or patients treated with long-term systemic steroidal therapy or other immunosuppressive therapies in the 14 days prior to enrollment.
10. Patients with other malignancies requiring progressive or aggressive treatment within the past 3 years.
11. Patients with active CNS metastases or carcinomatous meningitis.
12. Patients with severe (>= Grade 3) hypersensitivity to pembrolizumab or lenvatinib additives.
13. Patients with active autoimmune disease requiring systemic treatment, excluding replacement therapy within the past 2 years.
14. Patients with a history of interstitial lung disease/pneumonitis complicated by interstitial lung disease/pneumonitis or interstitial lung disease/pneumonitis requiring steroid administration (non-infectious).
15. Patients with active infections requiring systemic treatment at enrollment.
16. Patients with preexisting >=Grade 3 gastrointestinal or non-gastrointestinal fistula at enrollment.
17. Patients with gastrointestinal malabsorption or any other condition that might affect the absorption of lenvatinib.
18. Patients with a history of acute coronary syndrome, coronary angioplasty, or stenting within 6 months prior to enrollment.
19. Patients with clinically significant cardiovascular diseases within the 12 months prior to enrollment.
20. Patients having left ventricular ejection fraction (LVEF) below 50% within 14 days before the date of enrollment.
21. Patients with complications requiring hospitalization.
22. Patients with radiographic evidence of encasement or invasion of a major blood vessel, or of intratumoral cavitation.
23. Patients with QTcF spacing prolongation beyond 480 msec within 14 days prior to enrollment.
24. Patients with active hemoptysis in the 21 days prior to enrollment.
25. Patients who have comorbid psychotic or psychiatric symptoms or substance use disorders that interfere with daily life.
26. Patients with a history of allogeneic transplantation.